1. Field of the Invention
The present invention is related generally to an improved data processing system and in particular to a method and apparatus for generating cohort groups. More particularly, the present invention is directed to a computer implemented method, apparatus, and computer usable program code for optimizing selection of members of cohort groups across a plurality of dimensions.
2. Background Description
A cohort is a group of people or objects that share one or more attributes in common. An attribute is a characteristics or experience. For example, a group of people born in 1980 may form a birth cohort. A cohort may include one or more sub-cohorts. For example, the birth cohort of people born in 1980 may include a sub-cohort of people born in 1980 in Salt Lake City, Utah. A sub-sub-cohort may include people born in 1980 in Salt Lake City, Utah to low income, single parent households.
A cohort study is typically a longitudinal study that monitors or tracks cohort groups over time to identify trends, rates of disease in the cohorts, cohort behavior, success of medical treatments or pharmaceuticals, and/or other factors, events, and behaviors associated with the members of the cohort group. For a particular study to be successful, the members of the cohort groups may be required to possess one or more minimum attributes. The study may also be more effective if other attributes, such as life expectancy, age, and other attributes are also present. However, the number of possible attributes that may be of interest in a treatment study may be so numerous and/or complex that it is time consuming, expensive, difficult or impossible for a human user to optimize selection of members of cohort groups for the treatment study across all the attributes of interest.